ICDAR 2023 Competition on Indic Handwriting Text Recognition

Competition Updates

Registration Opens : January 01, 2023

Training Data Release : January 15, 2023

Validation Data Release : January 20, 2023

Test Data Release : March 10, 2023 March 20, 2023

Registration Close : March 14, 2023 April 2, 2023

Upload Brief Description of System/Algorithm/Network : March 15, 2023 April 10, 2023

Upload Results and Inference Code due : March 20, 2023 April 10, 2023

Winner Announcement : April 20, 2023

Recents Updates

Registration Opens : January 1, 2023

Training Data Release : January 15, 2023

Validation Data Release : January 20, 2023

Test Data Release : March 20, 2023

Leaderboard is up!!!

Winner Announcement : April 20, 2023


Optical Character Recognition (OCR) is a process of converting printed or handwritten images into machine-readable formats. OCR is an essential component in document image analysis. The ocr system usually consists of two main modules (i) text detection module and (ii) text recognition module. The text detection module locates all text blocks within an image, either at the word or line level. The text recognition module attempts to interpret the text image content and translate the visual signals into natural language tokens. The recognition of handwriting text is more challenging than the recognition of printed text for several reasons, including (i) diversity in handwriting styles, (ii) varying ink density around the words, (iii) challenging layouts with overlap between words and unstructured writing, and (iv) datasets with few writers and examples. These factors motivate and interest researchers to pursue research in such a challenging and complex field.

Many languages worldwide are disappearing due to their limited usage. We can easily use OCR and natural language processing techniques to stop the extermination of languages around the world. Among 7000 languages, the handwriting OCR is available only for a few languages such as English [1, 2, 3], Chinese [4, 5, 6], Arabic [7, 8], and Japanese [9, 10]. Among 22 languages in India, few are used only for communication purposes. Among these languages, Hindi, Bengali, and Telugu are the most spoken languages [11]. In most Indic scripts, two or more characters often combine to form conjunct characters [12]. These inherent features of Indic scripts make Handwriting Recognizer (HWR) more challenging than Latin scripts. Compared to the 52 unique (upper case and lower case) characters in English, most Indic scripts have over 100 unique basic Unicode characters [13]. Several Indian scripts and languages appear to be at risk due to the absence of research efforts. So, there is an immense need for research on text recognition for Indic scripts and languages.

Our challenge involves handwriting text recognition tasks in ten Indian scripts: Bengali, Devanagari, Gujarati, Gurumukhi, Kannada, Malayalam, Odia, Tamil, Telugu, and Urdu. This challenge aims to motivate researchers to develop/design new algorithms for the scripts mentioned above. We believe this challenge suggests a significant new direction for the community to recognize handwriting text in Indic scripts.


  1. Graves, A., Schmidhuber, J.: Offline handwriting recognition with multidimensional recurrent neural networks. In: NIPS (2008).
  2. Pham, V., Bluche, T., Kermorvant, C., Louradour, J.: Dropout improves recurrent neural networks for handwriting recognition. In: ICFHR (2014).
  3. Li, M., Lv, T., Cui, L., Lu, Y., Florencio, D., Zhang, C., Li, Z., Wei, F.: Trocr: Transformer-based optical character recognition with pre-trained models. arXiv (2021).
  4. Xie, Z., Sun, Z., Jin, L., Feng, Z., Zhang, S.: Fully convolutional recurrent network for handwritten Chinese text recognition. In: ICPR (2016)
  5. Wu, Y.C., Yin, F., Chen, Z., Liu, C.L.: Handwritten Chinese text recognition using separable multi-dimensional recurrent neural network. In: ICDAR (2017).
  6. Peng, D., Jin, L., Ma, W., Xie, C., Zhang, H., Zhu, S., Li, J.: Recognition of handwritten Chinese text by segmentation: A segment-annotation-free approach. IEEE Transactions on Multimedia (2022).
  7. Maalej, R., Kherallah, M.: Improving the DBLSTM for on-line Arabic handwriting recognition. Multimedia Tools and Applications (2020).
  8. Jemni, S.K., Ammar, S., Kessentini, Y.: Domain and writer adaptation of offline Arabic handwriting recognition using deep neural networks. Neural Computing and Applications (2022).
  9. Ly, N.T., Nguyen, C.T., Nakagawa, M.: Training an end-to-end model for offline handwritten Japanese text recognition by generated synthetic patterns. In: ICFHR (2018).
  10. Nguyen, K.C., Nguyen, C.T., Nakagawa, M.: A semantic segmentation-based method for handwritten Japanese text recognition. In: ICFHR (2020)
  11. Krishnan, P., Jawahar, C.V.: HWNet v2: An efficient word image representation for handwritten documents. IJDAR (2019).
  12. Script Grammar for Indian languages (Accessed March 26 2020), http://language.worldofcomputing.net/grammar/script-grammar.html.
  13. Pal, U., Chaudhuri, B.: Indian script character recognition: a survey. Pattern Recognition (2004).